Cisco Technology, Inc.

Stephen Williams

Rohit Bahl

Debashish Ghosh

Harsh Parandekar

Paul Clyde Sherrill

Various techniques are used to schedule computing jobs for execution by a computing resource. In an example method, a schedule is generated by selecting, for a first slot in the schedule, a first computing job based on a first priority of the first computing job with respect to a first characteristic. A second computing job is selected for a second slot in the schedule based on a second priority of the second computing job with respect to a second characteristic. The second slot occurs after the first slot in the schedule, and the second characteristic is different than the first characteristic. The first characteristic or the second characteristic includes an execution frequency. The computing jobs are executed based on the schedule.

Evolutionary modelling based non-disruptive scheduling and management of computation jobs

A multi-cloud service mesh orchestration platform can receive a request to deploy an application as a service mesh application. The platform can tag the application with governance information (e.g., TCO, SLA, provisioning, deployment, and operational criteria). The platform can partition the application into its constituent components, and tag each component with individual governance information. For first time steps, the platform can select and perform a first set of actions for deploying each component to obtain individual rewards, state transitions, and expected returns. The platform can determine a reinforcement learning policy for each component that maximizes a total reward for the application based on the individual rewards, state transitions, and expected returns of each first set of actions selected and performed for each component. For second time steps, the platform can select and perform a second set of actions for each component based on the reinforcement learning policy for the component.

Systems and methods for orchestrating microservice containers interconnected via a service mesh in a multi-cloud environment based on a reinforcement learning policy

In an embodiment, one or more non-transitory computer-readable storage media store one or more sequences of instructions, which when executed using one or more processors causes the one or more processors to perform various functions, such as accessing data stored in memory regarding a plurality of computation jobs. Such data includes, for instance, periodicity constraints that specify execution frequencies for the plurality of computation jobs, categorization data that categorizes the plurality of computation jobs into a plurality of job types, and organization data that organizes the plurality of computation jobs into a plurality of ordered arrangements. In this embodiment, there are at least a first ordered arrangement and a second ordered arrangement, each ordered arrangement comprises references to each of the plurality of computation jobs according to job type, and the job types are associated with relative priority indicia. The functions performed by the one or more processors also include populating a resulting ordered arrangement by selecting priority indicia from between respective job types in the first ordered arrangement and second ordered arrangement, and generating a schedule for execution of the plurality of computation jobs in accordance with the relative priority indicia in the resulting ordered arrangement and the periodicity constraints. The one or more processors are also configured to execute the one or more sequences of instructions to cause execution of the plurality of computation jobs according to the schedule.

In one embodiment, a system includes: a plurality of cloud nodes implemented on computing devices, the plurality of cloud nodes configured to execute computing jobs in a cloud computing environment according to a schedule, a schedule optimizer configured to: use a machine learning model to determine functional intent for job requests according to at least job execution metadata, and generate a schedule recommendation for jobs associated with the job requests, where the schedule recommendation is generated based at least in part on the functional intent; and a job executor configured to provide the schedule recommendation as the schedule to the plurality of cloud nodes, where the schedule optimizer and the job executor are instantiated in memory and executed by processing circuitry on at least one computing device.

Cloud resources optimization

In one embodiment, a method includes receiving at a network device comprising a data transfer optimization module, input identifying a source directory comprising data and a target directory at a storage device, splitting the data into a plurality of data groups for transfer to the storage device, transmitting the data groups concurrently to the storage device on data paths wherein transfer times of the data groups are monitored, receiving at the data transfer optimization module, identification of a data group with a longest transfer time, splitting the data group with the longest transfer time, and placing the data from the data group into at least two of the data groups for transfer to the storage device. An apparatus and logic are also disclosed herein.

Growing community of inventors

Stittsville, Canada

Stephen Williams